There are a number of risk factors for the development of DED (Fig. 11.1). Reference 1 lists these factors and stratifies them based on levels of evidence. The most obvious factors associated with DED are increasing age, female gender and hormonal changes. Connective tissue disease is associated with more severe forms of DED, e.g. Sjögren’s syndrome and graft-versus-host disease. Dry eye disease occurs with greater frequency in women, particularly in the menopausal and postmenopausal years. The preponderance of evidence has demonstrated that androgen insufficiency is a principal risk factor for both lacrimal gland involvement and meibomian gland dysfunction (MGD). The female : male ratio decreases with advancing age as large segments of males with MGD become affected. The distribution of subtypes of DED, based on glands primarily affected is discussed in the section on pathophysiology later.

Figure 11.1 Mechanisms of dry eye. The core mechanisms of dry eye are driven by tear hyperosmolarity and tear film instability. The cycle of events is shown on the right of the figure. Tear hyperosmolarity causes damage to the ocular surface epithelium by activating a cascade of inflammatory events at the ocular surface and a release of inflammatory mediators into the tears. Epithelial damage involves cell death by apoptosis, a loss of goblet cells and disturbance of the mucin expression. Individual etiologies often cause dry eye by several interacting mechanisms. ( (Text attenuated) Originally published in Report of the International Dry Eye Workshop (DEWS). The Ocular Surface 2007;5:85. Reproduced with permission of Elsevier, the publisher.)

Impact on Visual Function

A recent recognition is the impact that DED has on visual function. A hallmark of DED is tear film instability.² This is manifest in a rapid tear breakup time, during the inter-blink interval. This effect is not apparent in standard high-contrast visual acuity testing, in which the subject can blink and transiently re-form the tear film. However, within several seconds in dry eye patients, the tear film breaks up causing a significant loss of resolving power and a drop in acuity. A Japanese study has documented that inter-blink visual acuities of 20/60 and greater are common.^5,6 Patients rarely recognize the connection between their visual problem and dry eye, usually complaining of decreased reading time, or ocular fatigue. This is probably caused by repeated blinking in an unconscious effort to clear vision. Other parameters of visual loss in DED, such as contrast sensitivity and Shack Hartmann aberrometry measurements may be also affected.^7,8 These features of DED lead to the substantial degradation of quality of life measures in the disease. In one study, the equivalent burden of having moderate to severe dry eye was similar to that of moderate to severe angina.⁹

Perhaps the most significant recent clinical consequence of the visual changes has been reported in patients undergoing refractive surgery and cataract surgery with the implantation of high-resolution intraocular lenses. An unstable tear film interferes with the optical image sufficiently, such that the level of acuity expected is not achieved. Although large studies have yet to be published, recognition and identification of patients with DED preoperatively and medical management of the condition may improve visual outcomes.

Role of Symptoms in DED

The literature states that dry eye is a symptom-driven disease.¹ This implies that symptoms are a requisite component of the diagnosis. While symptoms certainly play a major role in the disease, recent studies have noted two seemingly paradoxical observations, i.e. that many patients with early/mild disease have symptoms greater than any objective findings, and that in a subset of patients with severe disease there is a paucity of symptoms. The latter is thought to be due to a decrease in sensory perception associated with inflammation; the former has been linked to a hyperalgesia, associated with early nerve ending response to injury. In a recent presentation of an observational study, only 60–70% of patients with DED were reported to have symptoms (as measured by the Ocular Surface Disease Index).¹⁰ The remaining 30–40% do not present with measurable symptoms. Reliance on symptoms alone for diagnosis would therefore, seem unwise.

Pathophysiology of Dry Eye Disease

Although there are numerous risk factors for the development of dry eye disease, the final common expression of the pathophysiology includes tear film hyperosmolarity and instability. The concept of a functional lacrimal unit has been developed in which the secretory glands (lacrimal, meibomian and mucin-producing surface cells), the entire ocular surface, and the lids are linked by a neural network, which responds to external stimuli to maintain a stable tear film and underlying ocular surface, necessary for subserving clear vision.¹¹ This unit has recently been enlarged to include the nasolacrimal duct, which has receptors that respond to volume changes.

A key concept to understanding the development of DED is the breakdown in this unit leading to an unstable, concentrated (hyperosmolar) tear film. In Table 11.1, this concept is illustrated in a graphic, which the entry points or risk factors in the periphery lead to the core central mechanisms.² As these mechanisms (e.g. lacrimal hyposecretion and/or meibomian gland dysfunction) come into play, there is a mutually reinforcing cycle of dysfunction leading to inflammation in the ocular surface and lacrimal glands. This vicious cycle concept has been developed by Baudouin¹² and is illustrated in Figure 11.1.

Table 11.1

Risk Factors for Dry Eye Disease

Mostly Consistent¹	Suggestive²	Unclear³
Older age	Asian race	Cigarette smoking
Female sex	Medications	Hispanic ethnicity
Postmenopausal estrogen therapy	Tricyclic antidepressants
Omega-3 and Omega-6 fatty acids	Selective serotonin reuptake inhibitors	Anticholinergics
Medications	Diuretics	Anxiolytics
Antihistamines	Beta-blockers	Antipsychotics
Connective tissue disease	Diabetes mellitus	Alcohol
LASIK and refractive excimer laser surgery	HIV/ HTLVl infection	Menopause
Radiation therapy	Systemic chemotherapy	Botulinum toxin injection
Hematopoietic stem cell transplantation	Large incision ECCE and penetrating keratoplasty
	Isotretinoin	Acne
Vitamin A deficiency	Low humidity environments	Gout
Hepatitis C infection	Sarcoidosis	Oral contraceptives
Androgen deficiency	Ovarian dysfunction	Pregnancy

¹ Mostly consistent evidence implies the existence of at least one adequately powered and otherwise well-conducted study published in a peer-reviewed journal, along with the existence of a plausible biological rationale and corroborating basic research or clinical data.

² Suggestive evidence implies the existence of either: (1) inconclusive information from peer-reviewed publications or (2) inconclusive or limited information to support the association, but either not published or published somewhere other than in a peer-reviewed journal.

³ Unclear evidence implies either directly conflicting information in peer-reviewed publications or inconclusive information but with some basis for a biological rationale.

(This table was published in Report of the International Dry Eye Workshop (DEWS) The Ocular Surface 2007;5:99. Reproduced with permission from Elsevier, the publisher.)

Principal Causative Factors

Tear hyperosmolarity is considered the central mechanism in DED, leading to inflammation at the ocular surface with resulting tissue damage and symptoms.² The hyperosmolar state is the result of either an insufficient secretion of fluid from the lacrimal glands (low aqueous flow) and/or excessive evaporation of the tear film. Normal tear osmolarity averages around 295 mOsmol/L which is isotonic with blood. Tear osmolarity over the ocular surface is greater than that in other tear compartments (e.g. the menisci), because of the larger surface area of the interpalpebral space, allowing for evaporative water loss between blinks. Indirect evidence of this comes from studies exposing the ocular surface in humans to increasing concentrations of solutions until ocular irritation occurs.¹³ It is also probable that, were it possible to sample tears from the preocular surface, the values would change over the ocular surface as the lid moves over the surface. There is a transfer of fluid from this preocular surface compartment to the tear menisci at their nasal extent with each complete blink. The menisci serve as a compartment of mixed tears and in normal subjects, is remarkably stable. As disease develops, the osmolarity in the tear menisci rises. While not the same as that over the ocular surface, there is a relationship that permits use of menisci to evaluate and monitor activity on the surface.

Hyperosmolarity affects the ocular surface in multiple ways. Hyperosmolarity initiates a cascade of inflammatory events mediated by at least two separate pathways, leading to the recruitment and activation of inflammatory cells on the ocular surface and the production of inflammatory cytokines and chemokines (see below). Associated with these changes is an increase in surface cell death (apoptosis), changes in mucin production and a loss of lubrication between the lid and the ocular surface.²

The lacrimal functional unit is capable of responding to these changes by increasing aqueous tear production in an attempt to lower the osmolarity. This may result in a transiently lower tear osmolarity but in aqueous tear deficient dry eye (ADDE). The lacrimal glands lack the secretory capacity to maintain this effort. In evaporative dry eye (EDE), most commonly caused by meibomian gland dysfunction (MGD), the excessive evaporative tear loss is associated with a thin or defective tear lipid layer which stimulates reflex tearing.¹⁴ Prolonged overstimulation of the lacrimal gland results in a neurogenic inflammation in the lacrimal gland, characterized by increased autoantigen expression, T-cell targeting and release of inflammatory cytokines into the tear film. These lead to compromise of the gland and a mixed or hybrid form of the disease. Other compensatory mechanisms may include increased blinking and squinting. Tear osmolarity has a linear relationship with increasing severity of dry eye¹⁰ and is the best most reliable metric for the diagnosis of DED.¹⁵

Tear Film Instability

The other hallmark of DED is tear film instability. In normal healthy subjects the lacrimal functional unit is remarkably stable in response to environmental stress. This is seen in tests of tear stability, such as the tear break-up time test (TBUT). With the onset of disease, an early feature is a breakdown in the homeostatic control of a stable tear film between blinks. This instability and tear hyperosmolarity are closely linked, but the primary event is unclear. Early studies suggested that a TBUT of less than 10 seconds was indicative of DED. Newer techniques to refine reproducibility and precision of the test include capture of the break-up process on video and the use of programmed videokeratography to decrease operator variability. Normative values between 5 and 7 seconds have been suggested.

The tear film should provide for a complete tear layer throughout the blink interval. When break-up of the tear film occurs prior to the next blink, there is a discontinuity that degrades the visual image and produces irritation. The ratio of the TBUT to the inter-blink interval has been called the ocular protection index.¹⁶ Values less than 1 are considered pathological.

Disturbances in the mucin cover of the cornea are responsible for rapid break-up. Tear break-up can occur over areas of corneal irregularity and are more rapid and extensive. Damage to the mucin layer or glycocalyx have been reported with the use of preservative-containing eye drops, particularly those containing benzalkonium chloride, which is a constituent of many glaucoma medications. Up to three-quarters of patients with long-term use of topical glaucoma medications exhibit symptoms and signs of DED, and the frequency increases with the number of medications used.¹⁷ Contact lens wear has also been associated with tear instability.

The Role of Inflammation

A major shift in our understanding of DED occurred with the recognition that inflammation plays a pivotal part in the pathogenesis of the disease. This is an important concept that has gained wide acceptance and is driving pharmaceutical development for more effective treatments. What follows is a summary of the evidence supporting the role of inflammation. For a more detailed discussion of this complex subject, readers are referred to the Cullen Symposium on Corneal and Ocular Surface Inflammation (Baylor College of Medicine, Houston, TX, January, 2005, The Ocular Surface, Vol. 3, Supplement). Patients with moderate to severe disease, usually show clinical evidence of inflammatory changes to the ocular surface and lids. In mild cases, these clinical signs are frequently absent.

Laboratory Studies

Both in vitro studies and animal models of dry eye disease have been utilized to study DED. Initial studies in canines demonstrated that topical treatment with cyclosporine A could effectively treat dry eye.¹⁸ The use of human conjunctival cell lines has demonstrated that inflammatory cytokines can up-regulate markers of inflammation, such as HLA-DR and ICAM-1. Studies utilizing mouse models, which usually involve initiation of disease by environmental stress, with or without systemic scopolamine injections, show inflammation of the conjunctiva and T-cell activation, which are seen in human DED.¹⁹ Figure 11.2 from Reference 19 illustrates the mechanisms thought to be involved in acute and chronic immune-associated inflammation in DED.

Figure 11.2 Hypothesis of the mechanism of acute and chronic immune inflammation. (Originally published in Report of the International Dry Eye Workshop (DEWS). The Ocular Surface 2007;5:185. Reproduced with permission of Elsevier, the publisher.)

Human Studies

Histopathological studies, measurement of inflammatory cytokines, chemokines and immunomodulatory molecules in tears, and tests for expression of cell surface antigens, all support the role of inflammation in human dry eye. Inflammation leads to tissue damage and is an intimate part of the vicious cycle of mutually reinforcing events in DED.² All of these markers of inflammation, which are seen in moderate to severe disease, are present to a lesser extent in mild to moderate cases. Many of the studies of inflammation in DED have employed patients suffering from Sjögren’s syndrome, graft-versus-host disease and other systemic conditions associated with autoimmune disease. In these cases, inflammation may play a primary role with autoimmune attack of the lacrimal gland as an early feature. In cases of dry eye associated with hormonal and aging changes, inflammation may play a more downstream role as a consequence of tear hyperosmolarity.

As mentioned above, the presence of tear hyperosmolarity has been shown to elicit a cascade of inflammatory events which involves both MAP kinases and NFκB signaling pathways from the ocular surface cells activating several pathways, leading to breakdown in the ocular surface and the presence of inflammatory cytokines, chemokines and tissue modulating markers in the tear film. Markers identified include IL, IL-1α,IL-17, TNF-α, and MMP-9. Inflammatory markers such as these and others are under study for their diagnostic possibilities.

Role of Androgens in DED

There is an extensive literature on the role of androgen deficiency in the pathogenesis of DED. Both low levels of systemic androgen and high levels of estrogen are risk factors for the development of DED. Biologically active androgens support the secretory activities of the lacrimal and meibomian glands. In addition, there are a number of associations between DED and other diseases characterized by decreased androgen activity, including complete androgen insensitivity syndrome, premature ovarian failure, and patients undergoing treatment with antiandrogen compounds for prostate cancer. A comprehensive review of this subject is published.²⁰

Distribution of Subtypes of Dry Eye Disease

As noted earlier, DED based on the glandular structures primarily involved has been divided into two mechanistic subtypes, i.e. aqueous tear deficient dry eye (ADDE) and evaporative dry eye (EDE). Although there are a number of conditions leading to the latter, the predominant factor is meibomian gland dysfunction (MGD). Aqueous deficiency has commonly been treated as synonymous with DED, with MGD occupying a position as a co-morbid condition. In recent years, there has been increased interest in the role of MGD in the pathogenesis of DED and its integral role in the lacrimal functional unit. Figure 11.3 from (Reference 2) depicts a comprehensive description of the mechanistic division of the different components of the ocular surface.

Figure 11.3 Major etiological causes of dry eye. (Originally published in Report of the International Dry Eye Workshop (DEWS). The Ocular Surface 2007;5:185. Reproduced with permission of Elsevier, the publisher.)

Studies in Asia, Europe, and more recently in the US, have documented the high prevalence of MGD, particularly in populations over the age of 50. MGD has become a major focus of research, and a comprehensive monograph on this subject was published in 2011. Early reports of the high prevalence of MGD in Asian populations led to the impression that MGD was more common in Asians.²¹ However, recent studies in Europe and the US have reported similarly high prevalences.^22,23 In the most recent report of 224 subjects, MGD alone far outweighs ADDE alone, 79 to 24, with 57 subjects showing evidence of a combined form of DED.²⁴ With increasing severity, there is a marked increase in the combined form of the disease, as initial compensatory mechanisms fail, compromising the other structures of the lacrimal functional unit. Most cases of Sjögren-associated DED are a mixed form of the disease. The recently published report of the MGD workshop thoroughly treats this topic.²⁵ A large population-based study in Spain has reported that more subjects with clear objective signs of MGD are asymptomatic than those with symptoms.²⁶